Exploratory Data Analysis

Dataset Summary

The datasets used in this project are a dataset of fires from MTBS and datasets of AQI from 2017 to 2021 from CBSA. The fires dataset contains information about 2,064 wildfires from 2017 to 2021. The AQI dataset has 97,243 AQI observations of states in the U.S. from 2017 to 2021. The average national AQIs for every year are 39.46, 38.87, 37.32, 36.74, 39.36, ordered from 2017 to 2021. The locations with the worst AQI every year were Country Of Mexico, Country Of Mexico, New Mexico, Oregon, South Dakota, ordered from 2017 to 2021. As a quick test to see if wildfires could be correlated with AQI, the highest AQI in Washington in 2021 was on 2021-08-13. The ignition day of the largest wildfire in Washington in 2021, on the other hand, was on 2021-07-07. It’s unknown how long the wildfire lasted with this dataset, but a one month difference should mean that there is a possibility that that wildfire increased the maximum AQI in Washington.

Table

Total acres burned every year in states
State_Abbr State 2017 2018 2019 2020 2021 Total_Burned
CA California 1296654 1579806 255278 4567669 262862 7962269
AK Alaska 704330 396222 2522419 176796 0 3799767
OR Oregon 724833 778701 61223 1065413 0 2630170
NV Nevada 1099528 932239 72780 259408 48828 2412783
AZ Arizona 379404 141978 348403 940504 3508 1813797
ID Idaho 664274 556852 263978 291052 18276 1794432
WA Washington 400869 404137 147970 764346 11622 1728944
MT Montana 1183049 89883 43309 305524 0 1621765
CO Colorado 97261 424061 32704 648440 0 1202466
UT Utah 228564 492880 79806 312244 9705 1123199
TX Texas 503074 310529 95958 146480 0 1056041
KS Kansas 700112 42629 20713 22773 0 786227
OK Oklahoma 245166 475227 21211 30213 0 771817
WY Wyoming 113701 216635 34040 262558 0 626934
NM New Mexico 120515 313923 66652 106910 0 608000
FL Florida 189940 145099 22965 54296 49629 461929
GA Georgia 166737 738 2277 0 771 170523
SD South Dakota 65062 0 0 9083 0 74145
HI Hawaii 1591 15617 13445 0 0 30653
MS Mississippi 10617 3936 0 10698 0 25251
MN Minnesota 0 17233 1293 1468 0 19994
LA Louisiana 503 2348 13717 1491 0 18059
NC North Carolina 8380 2400 0 2076 4533 17389
NE Nebraska 0 0 7587 9207 0 16794
NJ New Jersey 4189 827 8182 2952 0 16150
SC South Carolina 3025 3093 3406 0 1258 10782
MO Missouri 2234 3955 3276 1137 0 10602
KY Kentucky 2199 541 1900 0 3654 8294
MD Maryland 7743 0 0 0 0 7743
AR Arkansas 4543 2410 0 0 0 6953
VA Virginia 2309 2734 769 509 0 6321
AL Alabama 0 0 3618 941 624 5183
ND North Dakota 5056 0 0 0 0 5056
WV West Virginia 587 0 1327 0 0 1914
PA Pennsylvania 0 711 0 790 0 1501
TN Tennessee 1080 0 0 0 0 1080
MA Massachusetts 0 0 0 0 964 964
NY New York 0 658 0 0 0 658
CT Connecticut 0 0 0 0 0 0
DE Delaware 0 0 0 0 0 0
IL Illinois 0 0 0 0 0 0
IN Indiana 0 0 0 0 0 0
IA Iowa 0 0 0 0 0 0
ME Maine 0 0 0 0 0 0
MI Michigan 0 0 0 0 0 0
NH New Hampshire 0 0 0 0 0 0
OH Ohio 0 0 0 0 0 0
RI Rhode Island 0 0 0 0 0 0
VT Vermont 0 0 0 0 0 0
WI Wisconsin 0 0 0 0 0 0

For our table, we wanted to see how many acres were burned in every state, so it is sorted by the total amount of acres burned. To do this, we had to get all the wildfire data and the corresponding state within the timeline of 2017 to 2021. This table is included because we wanted to see which states consistently have wildfires, and which states do not. States on the western half of the United States seemed to fit this condition, while states on the eastern half had sporadic wildfires if at all. This allows us to determine how we should be comparing AQI. For states with consistently high amount of wildfires, it’ll be possible to compare AQI in fire seasons versus non fire seasons, as well as between years based on acres burned. For states with very few wildfires, it may be easier to just compare the AQI of years that did have wildfires with years that didn’t.

Chart 1: Choropleth

In order to see which states have had the most acres burned by wildfires, we decided to make a choropleth that maps each state with the total acres burned by wildfires from 2017 to 2021. This chart is color encoded by acres burned. White represents 0 acres burned while dark red represents 8 million acres, which is fitting since red is associated with fire. States can be hovered for detailed information such as state name and the totals for every year. As shown, California has the most acres burned by far, and generally the west coast seems to be more prone to wildfire burnage, including Alaska.

Chart 2: Scatter Plot

In order to see if there is a relationship between wildfires and AQI, we decided to make a scatter plot that graphs total burned acres with average AQI in 2021 for all states. A scatterplot was used since trends between two variables can be spotted. If there is a correlation, there should be some sort of line that shows up. Since not all states had wildfires in 2021, only the ones that had are graphed. Hovering over a point shows the state and the specific numbers. As shown, while there are strong cases like California with the worst AQI and most acres burned, most of the remaining states are fairly inconclusive.

Chart 3: Line

In order to see the AQI in Washington, we decided to make a line chart of the AQI each day in 2021. The use of this chart helps us see the pattern of AQI throughout the year in Washingtonand we can see that there was a large spike around the time of August and September when wildfireseason is the worst. Over the summer, there is drier heat, less rain, and more factors that contibuteto high levels of air quality. However, we can see that air quality remains consistent throughtout the rest of the year with no significantly noticable spikes.